Model comparison
DeepSeek-V3.1 vs Gemma 2 27B
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 29.4 on the Noometry Index.
Last verified . 20 shared benchmarks.
DeepSeek
GoogleSummary
- They share 20 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 7 categories and Gemma 2 27B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where DeepSeek-V3.1 leads 38.9 to 10.7.
- The biggest single-benchmark swing is DTBench: 82.7% for DeepSeek-V3.1 and 48% for Gemma 2 27B.
- DeepSeek-V3.1 is cheaper at $0.25 / $0.95 per million input/output tokens, against $0.65 / $0.65 for Gemma 2 27B.
- DeepSeek-V3.1 accepts more context: 164K tokens versus 8K.
Side by side
| DeepSeek-V3.1 | Gemma 2 27B | |
|---|---|---|
| Provider | DeepSeek | |
| Noometry Index | 42.8 | 29.4 |
| Released | 2025-08-21 | 2024-06-24 |
| Weights | Open | Open |
| Context window | 164K | 8K |
| Max output | 8K | 2K |
| Input $ / M tokens | $0.25 | $0.65 |
| Output $ / M tokens | $0.95 | $0.65 |
| Results tracked | 27 | 34 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Gemma 2 27B: 34.1 (#246)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Coding | 1417 | 1211 |
| WeirdML | 38.4% | — |
| BigCodeBench Instruct | — | 42.8% |
| LiveBench Coding | — | 36% |
| BigCodeBench Complete | — | 52.5% |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Gemma 2 27B: 15.3 (#315)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Hard Prompts | 1417 | 1198 |
| DTBench | 82.7% | 48% |
| LMCA | 24.3% | 7.1% |
| Epoch Capabilities Index | 139.92 | 122.08 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| LiveBench Reasoning | — | 28.1% |
| LiveBench Data Analysis | — | 47.9% |
| ForecastBench | 58 | — |
| LiveBench | — | 38.2% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Gemma 2 27B: 10.7 (#311)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Math | 1420 | 1212 |
| OTIS Mock AIME 2024-2025 | — | 1.4% |
| LiveBench Math | — | 26.5% |
| MATH Level 5 | — | 27.9% |
Knowledge DeepSeek-V3.1 leads
DeepSeek-V3.1: 43.7 (#90), Gemma 2 27B: 19.0 (#280)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Expert | 1405 | 1172 |
| GPQA Diamond | — | 36.5% |
| Confabulations | — | 27.1% |
| Vectara Hallucination Rate | 5.5% | — |
| MMLU | — | 75.7% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Gemma 2 27B: 38.6 (#226)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Non-English | 1400 | 1217 |
| LMArena Chinese | 1469 | 1221 |
| LMArena French | 1447 | 1247 |
| LMArena German | 1411 | 1209 |
| LMArena Japanese | 1378 | 1175 |
| LMArena Korean | 1337 | 1174 |
| LMArena Russian | 1405 | 1234 |
| LMArena Spanish | 1431 | 1228 |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Gemma 2 27B: 60.5 (#249)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Instruction Following | 1400 | 1206 |
| LiveBench Instruction Following | — | 58.1% |
Long Context Gemma 2 27B leads
DeepSeek-V3.1: 36.3 (#232), Gemma 2 27B: 37.3 (#218)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Longer Query | 1422 | 1231 |
| Fiction.LiveBench | 52.8% | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Gemma 2 27B: 44.2 (#225)
| Benchmark | DeepSeek-V3.1 | Gemma 2 27B |
|---|---|---|
| LMArena Text | 1420 | 1231 |
| LMArena Creative Writing | 1401 | 1241 |
| LMArena Multi-Turn | 1408 | 1224 |
| EQ-Bench Creative Writing | 1436 | — |
| LiveBench Language | — | 32.6% |
Frequently asked questions
Is DeepSeek-V3.1 better than Gemma 2 27B?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 29.4 on the Noometry Index.
Which is cheaper, DeepSeek-V3.1 or Gemma 2 27B?
DeepSeek-V3.1 is cheaper. It lists at $0.25 per million input tokens and $0.95 per million output tokens; Gemma 2 27B lists at $0.65 and $0.65.
Is DeepSeek-V3.1 or Gemma 2 27B better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 34.1 in the Noometry coding category.
Which has the bigger context window?
DeepSeek-V3.1 does, with 164K tokens against 8K.
How many benchmarks do DeepSeek-V3.1 and Gemma 2 27B share?
20 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Gemma 2 27B has 34.